Tom Effland

Thanks, but unless I'm mistaken, that's only for the training data links. I'm confused about how candidates are generated at _test_ time. There are many examples in the CoNLL testb...

Hmm, I'm confused then where the numbers for the WKD30, CoNLL, and TAC 2010 columns in Table 1.c in the paper are created from then... do you have an idea?

Hi, are there any plans in the future to support this? I really like dispatch, but this is a bit of an issue for downstream usability, since in python its...

@wesselb thanks for the detailed response. I've given it some consideration. I suppose one way to deal with this could be to take each function definition and convert it into...

Hi, Not sure how this compares to what you guys have been working on, but for what it's worth I have implemented a version of a biased rsample that uses...

Very interesting, I'll take a look at the paper -- unbiased forward sounds like a big plus. I'm available for a call to discuss pretty much whenever.

Quick dev question: when I try running `python setup.py test` or `python setup.py style`, I get `error: invalid command 'tests'` and `error: option --ignore not recognized`, respectively. I can run...

This appears to be something to do with the pytorch jit. E.g., [link](https://discuss.pytorch.org/t/torch-jit-trace-tracerwarning-converting-a-tensor-to-a-python-list-might-cause-the-trace-to-be-incorrect/26632). Are you using this with your model? As far as I can tell this is outside the...

Sure thing, I've replaced the colab link with the derivation and fixed the lint error (at least when I run it locally and drop the "tests" dir from the command)....

@srush Awesome, thanks for the fix. Sure, I can make these changes for the entropy, xent, and kl calculations. A quick question: is there a reason we call `self.logpartition()` when...